Artificial Intelligence (AI) has the potential to address systemic inequities and contribute to social good, yet marginalized communities—such as racial minorities and low-income populations—remain largely excluded from its benefits. Despite AI's promise in sectors like healthcare, climate resilience, and education, its deployment often reinforces biases and overlooks cultural and infrastructural challenges. This article addresses these issues by proposing a 10-step framework for developing culturally responsive, community-driven AI solutions that prioritize equity and inclusivity. By incorporating ethical, technical, and policy considerations, the framework aims to bridge the gap between AI advancements and the needs of underserved populations. The proposed approach has significant implications for fostering sustainable AI adoption that empowers marginalized communities and drives progress towards the United Nations’ Sustainable Development Goals (SDGs).